materials

Review

AC Corrosion of Carbon Steel under CathodicProtection Condition: Assessment, Criteria andMechanism. A Review

Andrea Brenna * , Silvia Beretta and Marco OrmelleseDipartimento di Chimica, Materiali e Ingegneria Chimica “Giulio Natta”, Politecnico di Milano, I-20131 Milan,Italy; silvia.beretta@polimi.it (S.B.); marco.ormellese@polimi.it (M.O.)* Correspondence: andrea.brenna@polimi.it

Received: 21 January 2020; Accepted: 6 May 2020; Published: 7 May 2020

Abstract: Cathodic protection (CP), in combination with an insulating coating, is a preventativesystem to control corrosion of buried carbon steel pipes. The corrosion protection of coating defectsis achieved by means of a cathodic polarization below the protection potential, namely −0.85 V vs.CSE (CSE, copper-copper sulfate reference electrode) for carbon steel in aerated soil. The presence ofalternating current (AC) interference, induced by high-voltage power lines (HVPL) or AC-electrifiedrailways, may represent a corrosion threat for coated carbon steel structures, although the potentialprotection criterion is matched. Nowadays, the protection criteria in the presence of AC, as well asAC corrosion mechanisms in CP condition, are still controversial and discussed. This paper dealswith a narrative literature review, which includes selected journal articles, conference proceedingsand grey literature, on the assessment, acceptable criteria and corrosion mechanism of carbon steelstructures in CP condition with AC interference. The study shows that the assessment of AC corrosionlikelihood should be based on the measurement of AC and DC (direct current) related parameters,namely AC voltage, AC and DC densities and potential measurements. Threshold values of thementioned parameters are discussed. Overprotection (EIR-free < −1.2 V vs. CSE) is the most dangerouscondition in the presence of AC: the combination of strong alkalization close to the coating defectdue to the high CP current density and the action of AC interference provokes localized corrosion ofcarbon steel.

Keywords: alternating current; cathodic protection; carbon steel; pipeline; AC interference corrosion;AC corrosion assessment; protection criteria; corrosion mechanism

1. Introduction

Cathodic protection (CP), in combination with an insulating coating, is a well-knownelectrochemical technique that reduces (or halts) the external corrosion rate of buried carbon steelpipes used to transport liquid or gas. In CP condition, the corrosion rate is reduced below 0.01 mm·a−1,which is the maximum acceptable value fixed by CP standards [1,2]. Carbon steel in aerated soil, i.e.,where oxygen reduction is the controlling cathodic process, operates in CP condition if the IR-freepotential (excluding the ohmic drop contribution in soil) is more negative than −0.85 V vs. CSE(Cu/CuSO4 reference electrode, +0.318 V vs. standard hydrogen electrode, SHE) [1,2].

The presence of alternating current (AC) interference on buried pipelines in free corrosion orunder CP condition can lead to severe localized corrosion through the pipe thickness. In the caseof AC interference, the sources of electrical disturbance are the high-voltage power lines (HVPL) orthe AC-electrified railways (fed by a high voltage line at 50 or 60 Hz), the receptor is the pipelinethat runs parallel to the interference source and the coupling mechanism occurs mainly via a resistive(or conductive) and inductive (or electromagnetic) mechanism [3].

Materials 2020, 13, 2158; doi:10.3390/ma13092158 www.mdpi.com/journal/materials

Materials 2020, 13, 2158 2 of 22

The resistive coupling is primarily a concern when there is a fault or an unbalanced conditionon the power line and large currents and voltages are conveyed to the earth during the HVPL shortcircuit. Although the interference time is short, it represents a hazard to the operators and to theburied pipe corresponding to the coating defects. The inductive coupling occurs when AC flowingin phase wires produces an electromagnetic field inducing alternating currents and voltages to thepipeline, which shares the way with the power line. The induced AC voltage depends on the length ofparallelism with the power line, and it is inversely proportional to the distance between the HVPL andthe pipeline; the AC density, i.e., the current for unit surface, is a function of AC voltage, the coatingdefect dimension and soil resistivity.

Nowadays, there is agreement that corrosion induced by AC interference can occur even oncarbon steel structures that fully respect the CP criterion and that AC corrosion is less than thatprovoked by the equivalent direct current, i.e., considering the same current density. In the presence ofAC interference, the CP criteria reported by international standards [1,2] are not sufficient to provethat steel is protected from corrosion. In the past 50 years, great effort has been made in order topropose criteria to assess AC corrosion likelihood and to understand the mechanism by which ACcauses corrosion. This effort brought about the international standard ISO 18086 (Corrosion of metalsand alloys—Determination of AC corrosion—Protection criteria) [4] that replaced in 2017 the EN15280 standard (Evaluation of a.c. corrosion likelihood of buried pipelines applicable to cathodicallyprotected pipelines). The ISO 18086 standard provides monitoring procedures, mitigation measuresand information to deal with long-term AC interference. Nevertheless, some aspects related to thephenomenon were not fully understood and the protection criteria as well as the corrosion mechanismhave been debated for a long time.

This paper deals with a narrative literature review, which includes a deep analysis of the ACcorrosion phenomenon, in particular the assessment of AC interference, and the evaluation of ACcorrosion likelihood and interference levels, the corrosion mechanism.

2. Assessment of AC Corrosion Likelihood

The assessment of AC interference likelihood on a buried pipeline should include severalparameters related to both the interference source and the interfered structure. During the design phase,the evaluation of AC interference on a buried structure can be carried out by mathematical/electricalmodelling, e.g., according to EN 50443 (Effects of electromagnetic interference on pipelines caused byhigh voltage AC electric traction systems and/or high voltage AC power supply systems) [5] or IEEEGuide for Safety in AC Substation Grounding [6]. These approaches aim to evaluate the tolerable ACvoltage based on parameters, such as the electrical configuration of the AC power line, the distancebetween the AC source (power line or traction system) and the pipeline, the insulation properties ofthe coating as well as soil resistivity. In the case of existing structures, field measurements can be usedas an alternative to calculation. According to calculations or field measurements, relevant mitigationmeasures should be installed to decrease the AC corrosion probability. Nevertheless, not only electricalparameters are involved in the AC corrosion mechanism and an electrochemical approach is requiredfor an understanding of the mechanism, in particular in the presence of cathodic protection.

According to ISO 18086 [4], the assessment of AC corrosion should be performed by evaluation ofsome or all of the following parameters:

• AC voltage, VAC;• AC density, iAC;• DC density, iDC;• AC/DC densities ratio, iAC/iDC;• DC potential (IR-free potential, EIR-free, and ON potential, EON);• Soil resistivity, ρ.

Materials 2020, 13, 2158 3 of 22

2.1. AC Voltage

The measurement of the AC voltage, VAC, on a pipeline is carried out with respect to a referenceelectrode located at remote position, i.e., where the AC voltage gradient does not change and is closeto zero. The AC voltage gradient is measured by means of two-reference electrodes spaced 1 to 5 mtransverse to the pipeline.

According to [4], acceptable AC voltages on the pipeline in CP condition are lower than 15 Vr.m.s. measured as an average over a representative time (e.g., 24 h). According to NACE SP0177(Mitigation of Alternating Current and Lightning Effects on Metallic Structures and Corrosion ControlSystems) [7], the maximum AC voltage is set at 15 V with respect to local earth (approximately 1 m);this threshold is mostly driven by safety considerations (shock hazard).

In a recent work, Tang et al. [8] investigate the effects of several parameters on the electricfield distribution of AC interference, such as the unbalanced current magnitude, soil and coatingresistivity and the distance between the power line and the pipeline. By means of numerical simulation,the authors conclude that the reference electrode should be placed farther from the pipeline route withthe increase of mitigation wire length, soil resistivity and the distance between the power line and thestructure; conversely, the earth remote position is closer to the pipe by increasing coating resistivity ifmitigation is applied [8].

2.2. AC Density

AC and DC densities on a coating defect control both the AC interference and the CP level,respectively. Contrary to the AC voltage measurement, the AC density, iAC, cannot be readilydetermined. The numerical approach considers the calculation of AC density from the AC voltage,the soil resistance, ρ, and the diameter, φ, of a circular coating defect, according to the followingequation, as reported in [8–12]:

iAC =8 ·VAC

ρ ·π ·φ(1)

Considering a maximum AC voltage of 15 V measured on a circular coating defect of 1 cm2

and a medium soil resistivity of 100 Ω·m, the expected AC density threshold is about 30 A·m−2.Nevertheless, this calculation is generally not possible since the coating defect area is not known.Moreover, the application of CP can significantly change the electrolyte composition in proximity tothe coating defect and consequently the local soil resistivity. The formula is valid when the coatingdefect size is larger than the coating thickness, although rigorous calculations are available [13].The current density can only be estimated by means of coupons or probes. According to ISO 18086 [4],the measurement of AC density has to be carried out on a 1 cm2 coupon surface area connected tothe structure.

The definition of a critical threshold of AC density over which AC corrosion could occur is stillcontroversial and large data variability is found. Compared to DC interference corrosion, AC corrosionof carbon steel is lower considering the same current density. Since the sixties of the last century [14,15],the effect of AC density was determined in terms of “equivalent DC density”, defined as the percentratio of the weight loss caused by AC to the expected weight loss due to the same DC density.The values of equivalent DC density are in the range between 0.1% and 0.3% with AC density up to600 A·m−2 [14,15]. Using laboratory tests on carbon steel in soil-simulating solution (1200 mg·dm−3

sulphates, 200 mg·dm−3 chlorides), Goidanich et al. [16] reported that AC corrosion efficiency (definedsimilarly to the “equivalent DC density”) is lower than 1% when AC density ranges from 50 to500 A·m−2, but it increases up to 4% for AC density lower than 50 A·m−2 (Figure 1). In 2010, Fu andCheng [17] reported comparable results.

Materials 2020, 13, 2158 4 of 22

Materials 2020, 13, x FOR PEER REVIEW 4 of 21

significantly increase in the presence of 20 A·m−2 AC density: the measured corrosion rate at 20 A·m−2 AC density is nearly two times higher than that of the control specimen in free corrosion condition and decreases with the application of CP. This last consideration introduces the need of the additional consideration of the cathodic DC density.

(a) (b)

Figure 1. Weight loss tests on carbon steel in free corrosion condition exposed to a soil-simulating solution: (a) corrosion rates vs. AC density (iAC) and (b) AC corrosion efficiency vs. AC density (iAC) as reported by Goidanich et al. [16].

2.3. AC/DC Current Density Ratio

For carbon steel structures under CP conditions, AC corrosion likelihood should be evaluated also considering the level of DC polarization, by means of the IR-free potential or DC density. The latter can be measured by means of a corrosion coupon or probes with a known surface area, e.g., 1 cm2. In order to assess AC corrosion conditions, it is better to refer to the AC density-DC density ratio (iAC/iDC), which is dimensionless. Nevertheless, use of only the iAC/iDC ratio could be misleading in the assessment of AC corrosion likelihood, i.e., different AC corrosion conditions can be represented by the same iAC/iDC ratio. For instance, an iAC/iDC ratio equals to 10 results from an interference condition of 30 A·m−2 AC density in the presence of 3 A·m−2 DC density or 3 A·m−2 AC density with 0.3 A·m−2 DC density. Although the ratio between the current densities is equal in the two conditions, they represent a different corrosion risk, i.e., 3 A·m−2 AC density is not recognized as a threat, dissimilarly from 30 A·m−2. As discussed in Paragraph 3, several authors [21–27] investigated the effect of iAC/iDC ratio on corrosion rate, proposing different threshold limits. ISO 18086 standard [4] reports that AC corrosion can be mitigated by maintaining the iAC/iDC ratio less than 3 over a representative time (e.g., 24 h) and it is valid for DC density greater than 10 A·m−2 (severe over-protection condition) and AC density over 30 A·m−2.

2.4. AC Frequency

There is full agreement that the corrosion rate decreases by increasing the frequency of the AC signal. The effect of frequency has been investigated on mild steel, nickel and copper-nickel alloys [28–35]. AC can cause severe corrosion at the industrial frequencies of 50 or 60 Hz, while the effect decreases at frequencies higher than 150 Hz.

Fernandes et al. [28] and the other authors [29–32] proposed a kinetic interpretation: by increasing the frequency, the time between the anodic and cathodic half-cycles becomes shorter and the metallic ions dissolved during the anodic period would be available for the subsequent deposition in the cathodic cycle.

Guo et al. [34,35] reported that at an AC density of 50 A·m−2, the corrosion rate of X60 steel decreases from 1.2 mm·a−1 at 10 Hz frequency to about 0.6 mm·a−1 at 50 Hz. In parallel, the free

Figure 1. Weight loss tests on carbon steel in free corrosion condition exposed to a soil-simulatingsolution: (a) corrosion rates vs. AC density (iAC) and (b) AC corrosion efficiency vs. AC density (iAC)as reported by Goidanich et al. [16].

Gummow et al. [18] stated that the corrosion rate increases with increased AC density greater than20 A·m−2 and becomes significant at AC densities greater than 100 A·m−2, regardless of the magnitudeof CP density. Based on laboratory tests, Pourbaix et al. [19] reported that AC corrosion is associatedwith the IR-free potential oscillation during interference but it is not related to a critical value of theAC density. As reported by Yunovich and Thompson [20], steel corrosion can significantly increase inthe presence of 20 A·m−2 AC density: the measured corrosion rate at 20 A·m−2 AC density is nearlytwo times higher than that of the control specimen in free corrosion condition and decreases with theapplication of CP. This last consideration introduces the need of the additional consideration of thecathodic DC density.

2.3. AC/DC Current Density Ratio

For carbon steel structures under CP conditions, AC corrosion likelihood should be evaluated alsoconsidering the level of DC polarization, by means of the IR-free potential or DC density. The lattercan be measured by means of a corrosion coupon or probes with a known surface area, e.g., 1 cm2.In order to assess AC corrosion conditions, it is better to refer to the AC density-DC density ratio(iAC/iDC), which is dimensionless. Nevertheless, use of only the iAC/iDC ratio could be misleading in theassessment of AC corrosion likelihood, i.e., different AC corrosion conditions can be represented by thesame iAC/iDC ratio. For instance, an iAC/iDC ratio equals to 10 results from an interference condition of30 A·m−2 AC density in the presence of 3 A·m−2 DC density or 3 A·m−2 AC density with 0.3 A·m−2 DCdensity. Although the ratio between the current densities is equal in the two conditions, they representa different corrosion risk, i.e., 3 A·m−2 AC density is not recognized as a threat, dissimilarly from30 A·m−2. As discussed in Paragraph 3, several authors [21–27] investigated the effect of iAC/iDC

ratio on corrosion rate, proposing different threshold limits. ISO 18086 standard [4] reports that ACcorrosion can be mitigated by maintaining the iAC/iDC ratio less than 3 over a representative time (e.g.,24 h) and it is valid for DC density greater than 10 A·m−2 (severe over-protection condition) and ACdensity over 30 A·m−2.

2.4. AC Frequency

There is full agreement that the corrosion rate decreases by increasing the frequency of theAC signal. The effect of frequency has been investigated on mild steel, nickel and copper-nickel

Materials 2020, 13, 2158 5 of 22

alloys [28–35]. AC can cause severe corrosion at the industrial frequencies of 50 or 60 Hz, while theeffect decreases at frequencies higher than 150 Hz.

Fernandes et al. [28] and the other authors [29–32] proposed a kinetic interpretation: by increasingthe frequency, the time between the anodic and cathodic half-cycles becomes shorter and the metallicions dissolved during the anodic period would be available for the subsequent deposition in thecathodic cycle.

Guo et al. [34,35] reported that at an AC density of 50 A·m−2, the corrosion rate of X60 steeldecreases from 1.2 mm·a−1 at 10 Hz frequency to about 0.6 mm·a−1 at 50 Hz. In parallel, the freecorrosion potential increases to about 50 mV. Yunovich and Thompson [36] proposed an electricalcircuit in order to simulate the behavior of a steel specimen exposed to soil varying the frequency ofthe AC signal. The model shows that the corrosion current in the circuit decreases with increasingfrequency and is approximately 0.3% of the total current at 60 Hz frequency, in agreement with theresults using weight loss tests, as reported in Figure 1.

2.5. Soil Resistivity and Chemical Composition

According to Equation (1), the AC density corresponding to a coating defect depends on thealternating voltage and on the spread resistance, which is the ohmic resistance through a coating defect(or a corrosion coupon) to earth. The ISO 18086 standard [4] reports an empirical relation between soilresistivity and AC corrosion risk:

• ρ < 25 Ω·m: very high risk;• 25 Ω·m < ρ < 100 Ω·m: high risk;• 100 Ω·m < ρ < 300 Ω·m: medium risk;• ρ > 300 Ω·m: low risk.

Soil resistivity close to a coating defect is significantly affected by the electrochemical reactions atthe metal-to-electrolyte interface, due to the application of the CP current. In CP condition, oxygenreduction (O2 + 2H2O + 4e−→ 4OH−) and, at lower potential, hydrogen evolution (2H+ + 2e−→ H2),cause a growth of alkalinity at the metal surface. The pH value can increase over 10 and up to 12–13at very high cathodic current densities. The local soil chemical composition can play a crucial rolein the AC corrosion assessment, as documented in [27,37]. Earth-alkaline ions (as Ca2+ and Mg2+),moved towards the metal surface by the CP electric field, form slightly soluble hydroxides; the pHincrease, shifting the carbonate-bicarbonate chemical equilibrium, favors the growth of a scale ofcalcium and magnesium carbonate that increases the spread resistance. Otherwise, alkaline cations (asNa+, K+ or Li+) form not-scaling hydroxides. Büchler et al. [37] reported a reduction of AC densitydue to the growth of chalk layers on the surface in the presence of calcium ions.

Recently, Xiao et al. [27] reported that the spread resistance of a X70 steel specimen at constantCP potential and different AC densities is higher in the presence of calcium and magnesium ions.Moreover, the corrosion rate of the specimens exposed to higher content of Na+ was greater than thatin the presence of earth-alkaline ions at the same potential and AC density (100 A·m−2, 300 A·m−2).

2.6. Effect on DC Potential (Free Corrosion Condition)

2.6.1. Negative Shift of Potential

There is general agreement that the free corrosion potential of carbon steel, i.e., without cathodicprotection, decreases as the AC density increases. This has been documented from the sixties of thelast century. Bolzoni et al. [38] reported laboratory tests on the influence of AC interference on carbonsteel corrosion in free corrosion condition in different environments (sulfate and chloride aqueoussolutions, with or without oxygen, simulating soil conditions and seawater). AC was overlapped tothe specimens ranging from 10 to 6000 A·m−2. The free corrosion potential of carbon steel in chlorideand sulfate solutions decreases as AC density increases. At AC densities below 100 A·m−2, the DC

Materials 2020, 13, 2158 6 of 22

potential variation was low (about 50 mV); above 100 A·m−2, the effect was higher (100–200 mV).In chloride solutions, the DC potential variation is less significant at high AC density (higher than1000 A·m−2). Results were confirmed in [39]: except for carbon steel in soil-simulating solution (1200ppm SO4

2− (Na2SO4) + 200 ppm Cl− (CaCl2·2H2O) [16]), the corrosion potential of galvanized steel,copper and carbon steel in different environmental conditions decreases with increasing AC density.Authors investigated the effects of AC density on anodic and cathodic overvoltages: AC has a significanteffect on the kinetics parameters, with a decrease of overvoltages and increase of exchange currentdensity of anodic and cathodic processes [39,40]. Nevertheless, in these papers some inconsistencieswere observed between the experimental tests and the results expected from mathematical modelsbased on the asymmetry of anodic and cathodic reactions. Li et al. [41] and Wang et al. [42] havemeasured a lowering of free corrosion potential on X70 and X80 carbon steel samples at variousAC densities in simulated soil solution and 3.5% sodium chloride solution. Potential shift in bothenvironments is about 0.2 V at 300 A·m−2 AC density. Moreover, authors investigated the kineticeffect of AC interference on anodic and cathodic overvoltages, measuring a variation of anodic andcathodic Tafel slope and their ratio in the presence of AC. Zhang et al. [43] proposed a nonlinear model(an electrical equivalent model considering the anodic and cathodic reactions under activation control)for the investigation of AC interference effect on corrosion potential and corrosion rate. Results showthat the expected variation of the free corrosion potential depends on the AC peak potential, as expected,and on the ratio of the anodic and cathodic Tafel slope (r = βa/βc). When r = 1 (symmetry of anodicand cathodic overvoltages), no DC potential variations are predicted by the model. For r > 1, a positive(anodic) potential shift is expected, while for r < 1 the DC potential lowers as AC peak potentialincreases. The latter covers the electrochemical condition of active carbon steel in soil or waters wherethe cathodic processes (oxygen reduction and/or hydrogen evolution) have a higher Tafel slope thanthat of steel dissolution. These data are consistent with the observations made in [44] and in previousworks [45–47].

2.6.2. Positive Shift of Potential

In 2012, He et al. [24] report that the average corrosion potential of X65 steel in loam soil moves tomore positive values by increasing AC density from 5 to 150 A·m−2. At 150 A·m−2, a positive shift ofabout 200 mV has been measured with respect to the condition without interference. Xu et al. [48]examined the effect of AC (60 Hz frequency) on 16Mn steel potential in a simulated soil solution bymeans of real-time AC/DC signal acquisition. AC moves corrosion potential negatively at an ACdensity lower than 400 A·m−2, while at higher AC density, the DC potential variation with respect to theabsence of AC interference is positive. In a recent work, Wu et al. [49] reported polarization curves ofX70 steel tested at AC densities up to 100 A·m−2 in simulated seawater. The presence of AC has a strongeffect on the polarization curves with a general shift toward higher current density and a positive(anodic) variation of the zero-current potential, i.e., the free corrosion potential. Nevertheless, as theAC density was raised from 10 to 100 A·m−2, the corrosion current density and the free corrosionpotential roughly remained constant.

2.7. Effect on DC Potential (Cathodic Protection Condition)

The potential measurement is affected by AC interference, even if CP is applied. Several authorshave investigated in the last decades the effect of AC on IR-free potential. Bolzoni et al. [38] investigatedthe influence of AC interference on carbon steel in CP condition in different environments (sulfateand chloride aqueous solutions, with or without oxygen). In the presence of cathodic polarization,the potential trend depends on DC density: at 0.1 A·m−2, the DC potential is lowered after ACapplication; conversely, at 1 and 10 A·m−2 DC density, the DC potential increases as the AC densityincreases. In 2008 [50], and later in 2010 [23], Ormellese et al. reported the measurements of IR-freepotential of carbon steel specimens exposed for about four months to a soil-simulating solution. DC andAC density were in the range 0.1–10 and 10–500 A·m−2, respectively. The increment of potential is not

Materials 2020, 13, 2158 7 of 22

significant at 10 A·m−2 AC, while it is about 0.1 and 0.2 V at 100 and 200 A·m−2, respectively. The effectof AC density on IR-free potential is more pronounced at high DC density [23].

Xu et al. [51,52] investigated the effects of AC on the CP potential reading of a 16Mn pipeline steelin a simulated soil solution. At −0.85 V vs. SCE (maintained in a galvanostatic way, SCE—saturatedcalomel electrode), AC moves DC potential negatively. Furthermore, the higher the AC density,the more negative the DC potential is. Conversely at −1 V vs. SCE, AC shifts potential in the positivedirection. Similar observations were reported by Kuang et al. [53,54]: the DC potential of X65 steel innear-neutral pH bicarbonate solution is shifted negatively by AC at −0.85 V vs. CSE, but positivelyshifted by AC under the CP of −1 V vs. CSE (Figure 2). Nevertheless, differently to what can beexpected, the potential variation reported is higher at smaller AC density (Figure 2b). When the appliedCP level was −0.925 V vs. CSE (data not shown), the DC potential becomes more positive at low ACdensities of 10 and 50 A·m−2, while it decreases with 100 A·m−2 AC density. Recently, Wang et al. [55]reported similar conclusions for X70 steel in near-neutral bicarbonate solution: at −0.775 V vs. SCE,the DC potential is shifted negatively consequently to the application of AC, while at −0.95 V vs. SCEand −1.2 V vs. SCE, an increase of DC potential is measured (Figure 3). In this case, the potentialvariation is proportional to AC density.

Materials 2020, 13, x FOR PEER REVIEW 7 of 21

et al. [55] reported similar conclusions for X70 steel in near-neutral bicarbonate solution: at −0.775 V vs. SCE, the DC potential is shifted negatively consequently to the application of AC, while at −0.95 V vs. SCE and −1.2 V vs. SCE, an increase of DC potential is measured (Figure 3). In this case, the potential variation is proportional to AC density.

Generally, it can be concluded that there is good agreement on the increase of DC potential in the presence of AC, although a negative shift is measured at small CP current density.

(a) (b)

Figure 2. Effect of AC density on DC potential of X65 steel in near-neutral pH bicarbonate solution in cathodic protection condition: (a) −0.85 V vs. CSE, (b) −1 V vs. CSE [54].

(a) (b)

Figure 3. Effect of AC density on DC potential of X70 steel in near-neutral pH bicarbonate solution in cathodic protection condition: (a) −0.775 V vs. saturated calomel electrode (SCE) (−0.850 V vs. CSE) (b) −0.95 V vs. SCE (−1.02 V vs. CSE). 1 mA·cm−2 corresponds to 10 A·m−2 [55].

3. Acceptable AC Interference Levels—Protection Criteria

There is large agreement that corrosion can occur on AC interfered carbon steel structures that fully match the CP potential criterion (E < Eprot) defined by ISO 15589-1 [2]. Much effort has been made in the last decades in order to define acceptable AC interference levels for carbon steel under CP condition. Kajiyama et al. [56–59] proposed a CP criterion based on the ratio between DC and AC densities, measured by means of corrosion coupons. The criterion can be summarized as follows (Figure 4): • if 0.1 A·m−2 ≤ iDC < 1 A·m−2, then iAC/iDC < 25, • if 1 A·m−2 ≤ iDC ≤ 20 A·m−2, then iAC < 70 A·m−2.

Accordingly, the maximum AC density depends on the CP level: at higher DC densities (i.e., more negative potential), a higher AC density can be tolerated. Even if the criterion has been applied successfully to some case studied [58], some authors recognized a greater AC corrosion risk at higher

Figure 2. Effect of AC density on DC potential of X65 steel in near-neutral pH bicarbonate solution incathodic protection condition: (a) −0.85 V vs. CSE, (b) −1 V vs. CSE [54].

Materials 2020, 13, x FOR PEER REVIEW 7 of 21

et al. [55] reported similar conclusions for X70 steel in near-neutral bicarbonate solution: at −0.775 V vs. SCE, the DC potential is shifted negatively consequently to the application of AC, while at −0.95 V vs. SCE and −1.2 V vs. SCE, an increase of DC potential is measured (Figure 3). In this case, the potential variation is proportional to AC density.

Generally, it can be concluded that there is good agreement on the increase of DC potential in the presence of AC, although a negative shift is measured at small CP current density.

(a) (b)

Figure 2. Effect of AC density on DC potential of X65 steel in near-neutral pH bicarbonate solution in cathodic protection condition: (a) −0.85 V vs. CSE, (b) −1 V vs. CSE [54].

(a) (b)

Figure 3. Effect of AC density on DC potential of X70 steel in near-neutral pH bicarbonate solution in cathodic protection condition: (a) −0.775 V vs. saturated calomel electrode (SCE) (−0.850 V vs. CSE) (b) −0.95 V vs. SCE (−1.02 V vs. CSE). 1 mA·cm−2 corresponds to 10 A·m−2 [55].

3. Acceptable AC Interference Levels—Protection Criteria

There is large agreement that corrosion can occur on AC interfered carbon steel structures that fully match the CP potential criterion (E < Eprot) defined by ISO 15589-1 [2]. Much effort has been made in the last decades in order to define acceptable AC interference levels for carbon steel under CP condition. Kajiyama et al. [56–59] proposed a CP criterion based on the ratio between DC and AC densities, measured by means of corrosion coupons. The criterion can be summarized as follows (Figure 4): • if 0.1 A·m−2 ≤ iDC < 1 A·m−2, then iAC/iDC < 25, • if 1 A·m−2 ≤ iDC ≤ 20 A·m−2, then iAC < 70 A·m−2.

Accordingly, the maximum AC density depends on the CP level: at higher DC densities (i.e., more negative potential), a higher AC density can be tolerated. Even if the criterion has been applied successfully to some case studied [58], some authors recognized a greater AC corrosion risk at higher

Figure 3. Effect of AC density on DC potential of X70 steel in near-neutral pH bicarbonate solution incathodic protection condition: (a) −0.775 V vs. saturated calomel electrode (SCE) (−0.850 V vs. CSE)(b) −0.95 V vs. SCE (−1.02 V vs. CSE). 1 mA·cm−2 corresponds to 10 A·m−2 [55].

Materials 2020, 13, 2158 8 of 22

Generally, it can be concluded that there is good agreement on the increase of DC potential in thepresence of AC, although a negative shift is measured at small CP current density.

3. Acceptable AC Interference Levels—Protection Criteria

There is large agreement that corrosion can occur on AC interfered carbon steel structures thatfully match the CP potential criterion (E < Eprot) defined by ISO 15589-1 [2]. Much effort has beenmade in the last decades in order to define acceptable AC interference levels for carbon steel underCP condition. Kajiyama et al. [56–59] proposed a CP criterion based on the ratio between DC andAC densities, measured by means of corrosion coupons. The criterion can be summarized as follows(Figure 4):

• if 0.1 A·m−2≤ iDC < 1 A·m−2, then iAC/iDC < 25,

• if 1 A·m−2≤ iDC ≤ 20 A·m−2, then iAC < 70 A·m−2.

Materials 2020, 13, x FOR PEER REVIEW 8 of 21

DC density differently to this criterion, as discussed later. Moreover, this criterion does not consider directly the value of the measured potential.

In 2012, He et al. [24] reported a similar approach based on current densities: the AC density threshold increases linearly with CP current density. The criterion (Figure 5) suggests there is not corrosion risk if iAC < 10 + 100·iDC (with iDC ≥ 0.01 A·m−2). Comparing the two criteria, the latter (Figure 5) is less conservative at a DC density lower than 1 A·m−2 and does not take into account AC corrosion at greater DC densities. For instance, at 0.1 A·m−2 DC density, the maximum allowed AC density is 2.5 and 20 A·m−2, considering the corrosion criterion of Figures 4 and 5, respectively.

Figure 4. Acceptable AC interference levels based on current densities criterion according to Kajiyama et al. (adapted from [56]).

Figure 5. Acceptable AC interference levels based on current densities criterion according to He et al. (adapted from [24]).

The effect of potential has been investigated by Ormellese et al. in [50] and later in [23,26]. The authors propose corrosion maps based on corrosion rate data evaluated using a weight loss test of carbon steel specimens exposed to a soil simulation environment, with varying AC interference and CP levels. Two AC corrosion risk regions are defined, low and high, for corrosion rates lower or greater than 10 μm·a−1, respectively. Corrosion risk increases by increasing the iAC/iDC ratio (Figure 6a): corrosion protection is achieved up to a maximum value of the iAC/iDC ratio, which decreases as the IR-free potential becomes more negative. Differently from the criteria discussed previously, in overprotection condition a few A·m−2 of AC density (ranging from 5 to 20 A·m−2, depending on potential) provokes corrosion of overprotected carbon steel. The authors proposed the following criterion (Figure 6b): • if 0.1 A·m−2 ≤ iDC < 1 A·m−2, then iAC < 30 A·m−2, • if 1 A·m−2 ≤ iDC ≤ 10 A·m−2, then iAC < 10 A·m−2.

The −0.85 V vs. CSE criterion is not always safe in the presence of AC interference; in overprotection condition (EIR-free < −1.2 V vs. CSE), severe AC corrosion could occur.

Fu et al. [60] proposed a criterion based on both potential and current densities: potentials more positive than −0.95 V vs. CSE are considered not safe in the presence of AC. The maximum acceptable

0.1; 2.5

1; 25

1; 70 20; 70

0.1

1

10

100

0.001 0.01 0.1 1 10 100

i AC

(A·m

-2)

iDC (A·m-2)

DC corrosion Protection

AC corrosion

Ove

rpro

tect

ion

0.01; 10

1; 110

0.1

1

10

100

0.001 0.01 0.1 1 10 100

i AC

(A·m

-2)

iDC (A·m-2)

Corrosion

Protection Possible overprotection

Figure 4. Acceptable AC interference levels based on current densities criterion according toKajiyama et al. (adapted from [56]).

Accordingly, the maximum AC density depends on the CP level: at higher DC densities (i.e.,more negative potential), a higher AC density can be tolerated. Even if the criterion has been appliedsuccessfully to some case studied [58], some authors recognized a greater AC corrosion risk at higherDC density differently to this criterion, as discussed later. Moreover, this criterion does not considerdirectly the value of the measured potential.

In 2012, He et al. [24] reported a similar approach based on current densities: the AC densitythreshold increases linearly with CP current density. The criterion (Figure 5) suggests there is notcorrosion risk if iAC < 10 + 100·iDC (with iDC ≥ 0.01 A·m−2). Comparing the two criteria, the latter(Figure 5) is less conservative at a DC density lower than 1 A·m−2 and does not take into account ACcorrosion at greater DC densities. For instance, at 0.1 A·m−2 DC density, the maximum allowed ACdensity is 2.5 and 20 A·m−2, considering the corrosion criterion of Figures 4 and 5, respectively.

Materials 2020, 13, 2158 9 of 22

Materials 2020, 13, x FOR PEER REVIEW 8 of 21

DC density differently to this criterion, as discussed later. Moreover, this criterion does not consider directly the value of the measured potential.

In 2012, He et al. [24] reported a similar approach based on current densities: the AC density threshold increases linearly with CP current density. The criterion (Figure 5) suggests there is not corrosion risk if iAC < 10 + 100·iDC (with iDC ≥ 0.01 A·m−2). Comparing the two criteria, the latter (Figure 5) is less conservative at a DC density lower than 1 A·m−2 and does not take into account AC corrosion at greater DC densities. For instance, at 0.1 A·m−2 DC density, the maximum allowed AC density is 2.5 and 20 A·m−2, considering the corrosion criterion of Figures 4 and 5, respectively.

Figure 4. Acceptable AC interference levels based on current densities criterion according to Kajiyama et al. (adapted from [56]).

Figure 5. Acceptable AC interference levels based on current densities criterion according to He et al. (adapted from [24]).

The effect of potential has been investigated by Ormellese et al. in [50] and later in [23,26]. The authors propose corrosion maps based on corrosion rate data evaluated using a weight loss test of carbon steel specimens exposed to a soil simulation environment, with varying AC interference and CP levels. Two AC corrosion risk regions are defined, low and high, for corrosion rates lower or greater than 10 μm·a−1, respectively. Corrosion risk increases by increasing the iAC/iDC ratio (Figure 6a): corrosion protection is achieved up to a maximum value of the iAC/iDC ratio, which decreases as the IR-free potential becomes more negative. Differently from the criteria discussed previously, in overprotection condition a few A·m−2 of AC density (ranging from 5 to 20 A·m−2, depending on potential) provokes corrosion of overprotected carbon steel. The authors proposed the following criterion (Figure 6b): • if 0.1 A·m−2 ≤ iDC < 1 A·m−2, then iAC < 30 A·m−2, • if 1 A·m−2 ≤ iDC ≤ 10 A·m−2, then iAC < 10 A·m−2.

The −0.85 V vs. CSE criterion is not always safe in the presence of AC interference; in overprotection condition (EIR-free < −1.2 V vs. CSE), severe AC corrosion could occur.

Fu et al. [60] proposed a criterion based on both potential and current densities: potentials more positive than −0.95 V vs. CSE are considered not safe in the presence of AC. The maximum acceptable

0.1; 2.5

1; 25

1; 70 20; 70

0.1

1

10

100

0.001 0.01 0.1 1 10 100

i AC

(A·m

-2)

iDC (A·m-2)

DC corrosion Protection

AC corrosion

Ove

rpro

tect

ion

0.01; 10

1; 110

0.1

1

10

100

0.001 0.01 0.1 1 10 100i A

C(A

·m-2

)

iDC (A·m-2)

Corrosion

Protection Possible overprotection

Figure 5. Acceptable AC interference levels based on current densities criterion according to He et al.(adapted from [24]).

The effect of potential has been investigated by Ormellese et al. in [50] and later in [23,26].The authors propose corrosion maps based on corrosion rate data evaluated using a weight loss test ofcarbon steel specimens exposed to a soil simulation environment, with varying AC interference and CPlevels. Two AC corrosion risk regions are defined, low and high, for corrosion rates lower or greater than10 µm·a−1, respectively. Corrosion risk increases by increasing the iAC/iDC ratio (Figure 6a): corrosionprotection is achieved up to a maximum value of the iAC/iDC ratio, which decreases as the IR-freepotential becomes more negative. Differently from the criteria discussed previously, in overprotectioncondition a few A·m−2 of AC density (ranging from 5 to 20 A·m−2, depending on potential) provokescorrosion of overprotected carbon steel. The authors proposed the following criterion (Figure 6b):

• if 0.1 A·m−2≤ iDC < 1 A·m−2, then iAC < 30 A·m−2,

• if 1 A·m−2≤ iDC ≤ 10 A·m−2, then iAC < 10 A·m−2.

Materials 2020, 13, x FOR PEER REVIEW 9 of 21

AC density increases as the potential becomes more negative: at −0.95 V vs. CSE, the maximum AC density is 20 A·m−2, while at −1.05 V vs. CSE, the threshold increases up to 100 A·m−2.

(a) (b)

Figure 6. AC corrosion protection criterion as proposed by Ormellese et al.: (a) AC corrosion risk diagram based on IR-free potential and current densities ratio, (b) protection criterion based on AC and DC (i.e., cathodic protection—CP) current densities.

Büchler in 2012 [61] and previously in 2009 [22] investigated new protection criteria based on laboratory and field investigations. For current density average values measured by means of on-site coupons, one of the criteria below must be met (Figure 7): • iAC < 30 A·m−2; • iDC < 1 A·m−2; • iAC/iDC < 3.

Nevertheless, the author [61] stated that the measurement with coupons is fraught with problems, since the obtained results are affected by coupon geometry and local soil conditions. The use of ON potential, EON, and AC voltage (VAC, indicated by the authors as UAC) is therefore suggested (Figure 8). For ON potential, one of the following criteria must be met: • average VAC < 15 V and average EON more positive than −1.2 V vs. CSE; • VAC < 3·(|EON| − 1.2) where EON is in V vs. CSE and EON < −1.2 V vs. CSE.

Figure 7. Protection criterion based on AC and DC densities as proposed by Büchler [61]. The region confined by the dotted lines corresponds to very severe AC corrosion and must be avoided.

-1.6

-1.5

-1.4

-1.3

-1.2

-1.1

-1.0

-0.9

-0.8

-0.7

1 10 100 1000 10000

EIR

-free

(V v

s. C

SE)

iAC/iDC

Serie1

Serie2

Serie3CR > 50 μm∙a-1

CR < 10 μm∙a-1

10 μm∙a-1 < CR < 50 μm∙a-1

Low AC corrosion risk

High AC corrosion risk

Figure 6. AC corrosion protection criterion as proposed by Ormellese et al.: (a) AC corrosion riskdiagram based on IR-free potential and current densities ratio, (b) protection criterion based on ACand DC (i.e., cathodic protection—CP) current densities.

The−0.85 V vs. CSE criterion is not always safe in the presence of AC interference; in overprotectioncondition (EIR-free < −1.2 V vs. CSE), severe AC corrosion could occur.

Fu et al. [60] proposed a criterion based on both potential and current densities: potentials morepositive than −0.95 V vs. CSE are considered not safe in the presence of AC. The maximum acceptableAC density increases as the potential becomes more negative: at −0.95 V vs. CSE, the maximum ACdensity is 20 A·m−2, while at −1.05 V vs. CSE, the threshold increases up to 100 A·m−2.

Büchler in 2012 [61] and previously in 2009 [22] investigated new protection criteria based onlaboratory and field investigations. For current density average values measured by means of on-sitecoupons, one of the criteria below must be met (Figure 7):

Materials 2020, 13, 2158 10 of 22

• iAC < 30 A·m−2;• iDC < 1 A·m−2;• iAC/iDC < 3.

Materials 2020, 13, x FOR PEER REVIEW 9 of 21

AC density increases as the potential becomes more negative: at −0.95 V vs. CSE, the maximum AC density is 20 A·m−2, while at −1.05 V vs. CSE, the threshold increases up to 100 A·m−2.

(a) (b)

Figure 6. AC corrosion protection criterion as proposed by Ormellese et al.: (a) AC corrosion risk diagram based on IR-free potential and current densities ratio, (b) protection criterion based on AC and DC (i.e., cathodic protection—CP) current densities.

Büchler in 2012 [61] and previously in 2009 [22] investigated new protection criteria based on laboratory and field investigations. For current density average values measured by means of on-site coupons, one of the criteria below must be met (Figure 7): • iAC < 30 A·m−2; • iDC < 1 A·m−2; • iAC/iDC < 3.

Nevertheless, the author [61] stated that the measurement with coupons is fraught with problems, since the obtained results are affected by coupon geometry and local soil conditions. The use of ON potential, EON, and AC voltage (VAC, indicated by the authors as UAC) is therefore suggested (Figure 8). For ON potential, one of the following criteria must be met: • average VAC < 15 V and average EON more positive than −1.2 V vs. CSE; • VAC < 3·(|EON| − 1.2) where EON is in V vs. CSE and EON < −1.2 V vs. CSE.

Figure 7. Protection criterion based on AC and DC densities as proposed by Büchler [61]. The region confined by the dotted lines corresponds to very severe AC corrosion and must be avoided.

-1.6

-1.5

-1.4

-1.3

-1.2

-1.1

-1.0

-0.9

-0.8

-0.7

1 10 100 1000 10000

EIR

-free

(V v

s. C

SE)

iAC/iDC

Serie1

Serie2

Serie3CR > 50 μm∙a-1

CR < 10 μm∙a-1

10 μm∙a-1 < CR < 50 μm∙a-1

Low AC corrosion risk

High AC corrosion risk

Figure 7. Protection criterion based on AC and DC densities as proposed by Büchler [61]. The regionconfined by the dotted lines corresponds to very severe AC corrosion and must be avoided.

Nevertheless, the author [61] stated that the measurement with coupons is fraught with problems,since the obtained results are affected by coupon geometry and local soil conditions. The use of ONpotential, EON, and AC voltage (VAC, indicated by the authors as UAC) is therefore suggested (Figure 8).For ON potential, one of the following criteria must be met:

• average VAC < 15 V and average EON more positive than −1.2 V vs. CSE;• VAC < 3·(|EON| − 1.2) where EON is in V vs. CSE and EON < −1.2 V vs. CSE.Materials 2020, 13, x FOR PEER REVIEW 10 of 21

Figure 8. Protection criterion based on AC voltage and ON potential, as proposed by Büchler [61]. The region below the dotted lines corresponds to acceptable AC interference conditions.

In 2015, these efforts brought to the AC corrosion protection criteria of the ISO 18086 standard [4] for buried carbon steel in CP condition:

• As a first step, the AC voltage on the pipeline should be decreased below 15 V r.m.s. The AC voltage is measured as an average over a representative time (e.g., 24 h) with respect to a reference electrode located in remote position;

• As a second step, AC corrosion mitigation is achieved by matching the CP protection potentials defined in ISO 15589-1 [2] and — maintaining the AC density (iAC) lower than 30 A·m−2 on a 1 cm2 coupon or probe over a representative time (e.g., 24 h), or — maintaining the average cathodic current density lower than 1 A·m−2 on a 1 cm2 coupon or probe over a representative time (e.g., 24 h), if AC density is higher than 30 A·m−2, or — maintaining the ratio between AC and DC densities (iAC/iDC) less than 3 over a representative time (e.g., 24 h).

In Annex E (informative), the standard reports other criteria that have been used in the presence of AC. These criteria have been derived from Büchler’s work [61] and are based on AC voltage, ON potential and current densities, as discussed. The protection criteria reported in the ISO 18086 standard are shown in Figure 9.

(a) (b)

Figure 9. Graphical representation of the AC protection criteria reported on ISO 18086 [4]: (a) iAC vs. iDC, (b) VAC vs. EON.

1; 30 10; 30

1

10

100

1000

0.001 0.01 0.1 1 10 100

i AC

(A·m

-2)

iDC (A·m-2)

Protection

Corrosion

0

2

4

6

8

10

12

14

16

18

20

-4 -3.5 -3 -2.5 -2 -1.5 -1

V AC

(V)

EON (V vs. CSE)

Corrosion

Protection

Prot

ectio

n

Figure 8. Protection criterion based on AC voltage and ON potential, as proposed by Büchler [61].The region below the dotted lines corresponds to acceptable AC interference conditions.

In 2015, these efforts brought to the AC corrosion protection criteria of the ISO 18086 standard [4]for buried carbon steel in CP condition:

Materials 2020, 13, 2158 11 of 22

• As a first step, the AC voltage on the pipeline should be decreased below 15 V r.m.s. The ACvoltage is measured as an average over a representative time (e.g., 24 h) with respect to a referenceelectrode located in remote position;

• As a second step, AC corrosion mitigation is achieved by matching the CP protection potentialsdefined in ISO 15589-1 [2] and — maintaining the AC density (iAC) lower than 30 A·m−2 on a 1 cm2

coupon or probe over a representative time (e.g., 24 h), or — maintaining the average cathodiccurrent density lower than 1 A·m−2 on a 1 cm2 coupon or probe over a representative time (e.g.,24 h), if AC density is higher than 30 A·m−2, or — maintaining the ratio between AC and DCdensities (iAC/iDC) less than 3 over a representative time (e.g., 24 h).

In Annex E (informative), the standard reports other criteria that have been used in the presenceof AC. These criteria have been derived from Büchler’s work [61] and are based on AC voltage,ON potential and current densities, as discussed. The protection criteria reported in the ISO 18086standard are shown in Figure 9.

Materials 2020, 13, x FOR PEER REVIEW 10 of 21

Figure 8. Protection criterion based on AC voltage and ON potential, as proposed by Büchler [61]. The region below the dotted lines corresponds to acceptable AC interference conditions.

In 2015, these efforts brought to the AC corrosion protection criteria of the ISO 18086 standard [4] for buried carbon steel in CP condition:

• As a first step, the AC voltage on the pipeline should be decreased below 15 V r.m.s. The AC voltage is measured as an average over a representative time (e.g., 24 h) with respect to a reference electrode located in remote position;

• As a second step, AC corrosion mitigation is achieved by matching the CP protection potentials defined in ISO 15589-1 [2] and — maintaining the AC density (iAC) lower than 30 A·m−2 on a 1 cm2 coupon or probe over a representative time (e.g., 24 h), or — maintaining the average cathodic current density lower than 1 A·m−2 on a 1 cm2 coupon or probe over a representative time (e.g., 24 h), if AC density is higher than 30 A·m−2, or — maintaining the ratio between AC and DC densities (iAC/iDC) less than 3 over a representative time (e.g., 24 h).

In Annex E (informative), the standard reports other criteria that have been used in the presence of AC. These criteria have been derived from Büchler’s work [61] and are based on AC voltage, ON potential and current densities, as discussed. The protection criteria reported in the ISO 18086 standard are shown in Figure 9.

(a) (b)

Figure 9. Graphical representation of the AC protection criteria reported on ISO 18086 [4]: (a) iAC vs. iDC, (b) VAC vs. EON.

1; 30 10; 30

1

10

100

1000

0.001 0.01 0.1 1 10 100

i AC

(A·m

-2)

iDC (A·m-2)

Protection

Corrosion

0

2

4

6

8

10

12

14

16

18

20

-4 -3.5 -3 -2.5 -2 -1.5 -1

V AC

(V)

EON (V vs. CSE)

Corrosion

Protection

Prot

ectio

n

Figure 9. Graphical representation of the AC protection criteria reported on ISO 18086 [4]: (a) iAC vs.iDC, (b) VAC vs. EON.

In 2018, Junker et al. [62] reported results of laboratory and field tests varying AC and DC levelsand soil chemistry. Both laboratory and field data confirmed very high AC corrosion rates underexcessive CP (> 1 A·m−2) and AC interference higher than 30 A·m−2. Moreover, they recognized inthe spread resistance of a coating defect a highly dynamic parameter under AC and DC influence.The investigations illustrate that the chemical environment alters the AC and DC density limits for ACcorrosion, however the present limits of ISO 18086 constitute a safe strategy in most environments.

Figure 10 reports the AC corrosion rate measured on corrosion coupons with varying AC andDC density, as reported by Nielsen [63]. Data are overlapped to the AC protection criteria reported inISO 18086.

Materials 2020, 13, 2158 12 of 22

Materials 2020, 13, x FOR PEER REVIEW 11 of 21

In 2018, Junker et al. [62] reported results of laboratory and field tests varying AC and DC levels and soil chemistry. Both laboratory and field data confirmed very high AC corrosion rates under excessive CP (> 1 A·m−2) and AC interference higher than 30 A·m−2. Moreover, they recognized in the spread resistance of a coating defect a highly dynamic parameter under AC and DC influence. The investigations illustrate that the chemical environment alters the AC and DC density limits for AC corrosion, however the present limits of ISO 18086 constitute a safe strategy in most environments.

Figure 10 reports the AC corrosion rate measured on corrosion coupons with varying AC and DC density, as reported by Nielsen [63]. Data are overlapped to the AC protection criteria reported in ISO 18086.

Figure 10. AC corrosion rate measured on corrosion coupons as reported by Nielsen [63]. Data are compared with the AC protection criteria reported in ISO 18086.

It can be concluded that overprotection (namely EIR-free lowers than −1.2 V vs. CSE [2] or iDC > 1 A·m−2) is the most dangerous condition in the presence of AC interference. At “high” CP levels, the maximum tolerable AC density is 30 A·m−2. Below 1 A·m−2 DC density, the AC corrosion likelihood decreases. Nevertheless, some doubts are revealed regarding the inexistence of an AC density threshold at “low” CP condition (iDC < 1 A·m−2).

4. AC Corrosion Mechanism

Numerous theories and models have been proposed on the AC corrosion mechanism of carbon steel. Some models referring to carbon steel structures in free corrosion conditions as well as in the presence of CP are discussed hereafter.

4.1. Effect of Anodic and Cathodic AC Half-Wave on Metal Dissolution

Büchler et al. [22,61] proposed a corrosion mechanism based on thermodynamic and kinetic considerations on the reactions involved during AC interference on cathodically protected carbon steel. When an AC voltage is present, current will flow through the coating defects exposed to soil. If the pH value is sufficiently high (above 10, as in CP condition), during the anodic half-cycle, steel oxidation occurs, promoting the formation of a passive film. During the cathodic half-wave, the passive film is electrochemically destroyed and converted in porous rust. In the successive anodic cycle, the passive film is reformed under the non-protective rust layer. Moreover, the Fe2+ present in the rust layer is oxidized to Fe3+ (Fe2+ → Fe3+ + e−). In the subsequent cathodic cycle, the dissolution of the passive film will increase the volume of porous rust. Hence, every AC cycle results in the oxidation of the metal with a significant metal loss in the long term (Figure 11). A simplified description of this mechanism is reported also in the ISO 18086 standard [4].

Figure 10. AC corrosion rate measured on corrosion coupons as reported by Nielsen [63]. Data arecompared with the AC protection criteria reported in ISO 18086.

It can be concluded that overprotection (namely EIR-free lowers than −1.2 V vs. CSE [2] or iDC >

1 A·m−2) is the most dangerous condition in the presence of AC interference. At “high” CP levels,the maximum tolerable AC density is 30 A·m−2. Below 1 A·m−2 DC density, the AC corrosion likelihooddecreases. Nevertheless, some doubts are revealed regarding the inexistence of an AC density thresholdat “low” CP condition (iDC < 1 A·m−2).

4. AC Corrosion Mechanism

Numerous theories and models have been proposed on the AC corrosion mechanism of carbonsteel. Some models referring to carbon steel structures in free corrosion conditions as well as in thepresence of CP are discussed hereafter.

4.1. Effect of Anodic and Cathodic AC Half-Wave on Metal Dissolution

Büchler et al. [22,61] proposed a corrosion mechanism based on thermodynamic and kineticconsiderations on the reactions involved during AC interference on cathodically protected carbon steel.When an AC voltage is present, current will flow through the coating defects exposed to soil. If the pHvalue is sufficiently high (above 10, as in CP condition), during the anodic half-cycle, steel oxidationoccurs, promoting the formation of a passive film. During the cathodic half-wave, the passive film iselectrochemically destroyed and converted in porous rust. In the successive anodic cycle, the passivefilm is reformed under the non-protective rust layer. Moreover, the Fe2+ present in the rust layer isoxidized to Fe3+ (Fe2+

→ Fe3+ + e−). In the subsequent cathodic cycle, the dissolution of the passivefilm will increase the volume of porous rust. Hence, every AC cycle results in the oxidation of the metalwith a significant metal loss in the long term (Figure 11). A simplified description of this mechanism isreported also in the ISO 18086 standard [4].

Materials 2020, 13, 2158 13 of 22Materials 2020, 13, x FOR PEER REVIEW 12 of 21

Figure 11. Schematic representation of AC corrosion mechanism according to Büchler et al. [61].

4.2. The Alkalization Mechanism and the Effect of Spread Resistance

Nielsen et al. [64–66] proposed the “alkalization model” of AC-induced corrosion of carbon steel under CP condition. The model is based on the combination of two effects: (1) the alkalization of the metal-to-electrolyte interface of overprotected carbon steel, and (2) potential oscillations across the immunity, the passive and the high-pH corrosion domain of the iron potential-pH diagram.

As known, the presence of a cathodic current on the metal surface under CP condition is beneficial because it promotes the reduction (or zeroing) of the corrosion rate and the increase of alkalinity due to the accumulation of hydroxyl ions (OH−) close to coating defect. The pH increase is proportional to the cathodic current density and depends on the diffusion and electrical migration of ions towards and from the metal surface. In overprotection condition, namely IR-free potential more negative than −1.2 V vs. CSE, the high cathodic current density (in the order of a few A·m−2) can promote a significant increase of pH up to 13 or higher. According to the Pourbaix diagram, at elevated pH the potential oscillations caused by AC interference could lead to periodic entry in the high-pH corrosion region with formation of dissolved HFeO2− ions. The authors report the presence of an “incubation time” defined as the period to reach a critical pH (close to 14) at the metal-to-soil interface, with a significant lowering of the spread resistance and increase of AC density due to depolarization effects of the AC (Figure 12). Then, potential oscillations could cause corrosion due to different time constants associated to iron dissolution (fast) and the formation of a passive film (slower). Accordingly, AC corrosion of carbon steel in CP condition cannot be reduced by adding a surplus of cathodic current, as in the case of DC corrosion phenomena, but by avoiding high DC densities and the overprotection condition, in agreement with the protection criteria of the ISO standard [4].

Figure 12. Schematic representation of the AC corrosion mechanism of carbon steel under CP condition, as proposed by Nielsen et al. [63–66].

Pipe wall

Coating CoatingCoating defect

Induced AC

Small coating defect

High CP current densityReduction spread resistance

High AC densityAC corrosion

Alkalinity

Figure 11. Schematic representation of AC corrosion mechanism according to Büchler et al. [61].

4.2. The Alkalization Mechanism and the Effect of Spread Resistance

Nielsen et al. [64–66] proposed the “alkalization model” of AC-induced corrosion of carbon steelunder CP condition. The model is based on the combination of two effects: (1) the alkalization of themetal-to-electrolyte interface of overprotected carbon steel, and (2) potential oscillations across theimmunity, the passive and the high-pH corrosion domain of the iron potential-pH diagram.

As known, the presence of a cathodic current on the metal surface under CP condition is beneficialbecause it promotes the reduction (or zeroing) of the corrosion rate and the increase of alkalinity dueto the accumulation of hydroxyl ions (OH−) close to coating defect. The pH increase is proportional tothe cathodic current density and depends on the diffusion and electrical migration of ions towardsand from the metal surface. In overprotection condition, namely IR-free potential more negative than−1.2 V vs. CSE, the high cathodic current density (in the order of a few A·m−2) can promote a significantincrease of pH up to 13 or higher. According to the Pourbaix diagram, at elevated pH the potentialoscillations caused by AC interference could lead to periodic entry in the high-pH corrosion regionwith formation of dissolved HFeO2

− ions. The authors report the presence of an “incubation time”defined as the period to reach a critical pH (close to 14) at the metal-to-soil interface, with a significantlowering of the spread resistance and increase of AC density due to depolarization effects of the AC(Figure 12). Then, potential oscillations could cause corrosion due to different time constants associatedto iron dissolution (fast) and the formation of a passive film (slower). Accordingly, AC corrosion ofcarbon steel in CP condition cannot be reduced by adding a surplus of cathodic current, as in the caseof DC corrosion phenomena, but by avoiding high DC densities and the overprotection condition,in agreement with the protection criteria of the ISO standard [4].

Materials 2020, 13, x FOR PEER REVIEW 12 of 21

Figure 11. Schematic representation of AC corrosion mechanism according to Büchler et al. [61].

4.2. The Alkalization Mechanism and the Effect of Spread Resistance

Nielsen et al. [64–66] proposed the “alkalization model” of AC-induced corrosion of carbon steel under CP condition. The model is based on the combination of two effects: (1) the alkalization of the metal-to-electrolyte interface of overprotected carbon steel, and (2) potential oscillations across the immunity, the passive and the high-pH corrosion domain of the iron potential-pH diagram.

As known, the presence of a cathodic current on the metal surface under CP condition is beneficial because it promotes the reduction (or zeroing) of the corrosion rate and the increase of alkalinity due to the accumulation of hydroxyl ions (OH−) close to coating defect. The pH increase is proportional to the cathodic current density and depends on the diffusion and electrical migration of ions towards and from the metal surface. In overprotection condition, namely IR-free potential more negative than −1.2 V vs. CSE, the high cathodic current density (in the order of a few A·m−2) can promote a significant increase of pH up to 13 or higher. According to the Pourbaix diagram, at elevated pH the potential oscillations caused by AC interference could lead to periodic entry in the high-pH corrosion region with formation of dissolved HFeO2− ions. The authors report the presence of an “incubation time” defined as the period to reach a critical pH (close to 14) at the metal-to-soil interface, with a significant lowering of the spread resistance and increase of AC density due to depolarization effects of the AC (Figure 12). Then, potential oscillations could cause corrosion due to different time constants associated to iron dissolution (fast) and the formation of a passive film (slower). Accordingly, AC corrosion of carbon steel in CP condition cannot be reduced by adding a surplus of cathodic current, as in the case of DC corrosion phenomena, but by avoiding high DC densities and the overprotection condition, in agreement with the protection criteria of the ISO standard [4].

Figure 12. Schematic representation of the AC corrosion mechanism of carbon steel under CP condition, as proposed by Nielsen et al. [63–66].

Pipe wall

Coating CoatingCoating defect

Induced AC

Small coating defect

High CP current densityReduction spread resistance

High AC densityAC corrosion

Alkalinity

Figure 12. Schematic representation of the AC corrosion mechanism of carbon steel under CP condition,as proposed by Nielsen et al. [63–66].

This “vicious circle” is supported by the data shown in Figure 13a–f [63], which illustratesexperimental results in a laboratory soil box environment in purified quartz sand. At a constant

Materials 2020, 13, 2158 14 of 22

AC voltage (15 V), six different levels of CP (ON potential) were applied for some weeks to monitorthe corrosion rate of an electrical resistance probe and various electrical parameters. At a fixed ACvoltage, the corrosion rate depends strongly on the CP level (ON potential), and therefore, AC voltagealone cannot be considered a valid indicator of AC corrosion risk. Despite the constant AC voltage,the AC density varies from about 100 A·m−2 at low CP levels up to 500 A·m−2 at higher CP levels.Figure 13e shows corrosion rate as a function of cathodic current density: the corrosion rate increaseswith increasing CP current density, in agreement with the proposed mechanism. At the same time,the spread resistance is strongly influenced by the DC density with the consequent increase of ACdensity and corrosion rate (Figure 13f).

Materials 2020, 13, x FOR PEER REVIEW 13 of 21

This “vicious circle” is supported by the data shown in Figure 13a–f [63], which illustrates experimental results in a laboratory soil box environment in purified quartz sand. At a constant AC voltage (15 V), six different levels of CP (ON potential) were applied for some weeks to monitor the corrosion rate of an electrical resistance probe and various electrical parameters. At a fixed AC voltage, the corrosion rate depends strongly on the CP level (ON potential), and therefore, AC voltage alone cannot be considered a valid indicator of AC corrosion risk. Despite the constant AC voltage, the AC density varies from about 100 A·m−2 at low CP levels up to 500 A·m−2 at higher CP levels. Figure 13e shows corrosion rate as a function of cathodic current density: the corrosion rate increases with increasing CP current density, in agreement with the proposed mechanism. At the same time, the spread resistance is strongly influenced by the DC density with the consequent increase of AC density and corrosion rate (Figure 13f).

The crucial role of soil chemical composition, pH and spread resistance was carefully investigated by Junker et al. [62,67–69]. The spread resistance is identified as a key parameter, controlling the current densities, and is highly influenced by the formation of calcareous deposits or corrosion products. At high CP density, AC corrosion of an ER probe element is associated with strong alkalization of the electrolyte and consequently dense calcareous deposit formation. The calcareous deposit dramatically increases the spread resistance and reduces the AC and DC densities. Corrosion decreases, but only as long as the calcareous deposit is stable and fully covering the surface. Due to brittle fracture or a ‘flake of’ mechanism of the scale (probably provoked by hydrogen evolution at low potentials), the spread resistance suddenly decreases, causing an increase of current densities and AC corrosion. The cathodic reactions on the re-exposed probe surface will restart the alkalization and precipitation of calcareous deposits and with time (days) the corrosion stops again. This causes a cyclic variation of spread resistance, current densities and corrosion rate. The detailed chemical investigation of stone hard soil formed on cathodically protected pipeline under AC interference is reported in [68].

Figure 13. Example of corrosion rate vs. various electrical parameters, as reported by Nielsen [63]. Figure 13. Example of corrosion rate vs. various electrical parameters, as reported by Nielsen [63].

The crucial role of soil chemical composition, pH and spread resistance was carefully investigatedby Junker et al. [62,67–69]. The spread resistance is identified as a key parameter, controlling the currentdensities, and is highly influenced by the formation of calcareous deposits or corrosion products.At high CP density, AC corrosion of an ER probe element is associated with strong alkalization of theelectrolyte and consequently dense calcareous deposit formation. The calcareous deposit dramaticallyincreases the spread resistance and reduces the AC and DC densities. Corrosion decreases, but onlyas long as the calcareous deposit is stable and fully covering the surface. Due to brittle fracture ora ‘flake of’ mechanism of the scale (probably provoked by hydrogen evolution at low potentials),the spread resistance suddenly decreases, causing an increase of current densities and AC corrosion.The cathodic reactions on the re-exposed probe surface will restart the alkalization and precipitation ofcalcareous deposits and with time (days) the corrosion stops again. This causes a cyclic variation ofspread resistance, current densities and corrosion rate. The detailed chemical investigation of stonehard soil formed on cathodically protected pipeline under AC interference is reported in [68].

The effect of spread resistance on AC corrosion has been also documented by Nielsen andCohn [70] with the help of an electrical equivalent circuit analysis that represents the impedances

Materials 2020, 13, 2158 15 of 22

existing between the pipe and remote earth. AC and DC sources impose a DC and AC voltage onthe pipe: the simulated AC source is the HVTL (High Voltage Transmission Line), whereas the DCsource represents the CP system. The authors consider static and dynamic elements. Static elements,namely spread resistance and charge transfer resistance, are defined as elements where impedance isfrequency independent. Conversely, dynamic elements are frequency dependent: these include thedouble layer capacitance and diffusion elements. Because of the low impedance of the capacitance,the spread resistance is the dominant impedance element at 50–60 Hz frequency and plays a key rolein controlling the AC corrosion process.

4.3. Effect of AC on Anodic and Cathodic Overvoltages

The increase of corrosion rate in the presence of AC has been explained by some authors using theeffect of AC on anodic and cathodic overvoltages [39,46,49,65,71,72]. Goidanich et al. [39] investigatedby means of laboratory tests the influence of AC on kinetic characteristics of carbon steel, galvanizedsteel, copper and zinc under different experimental conditions. Results showed that AC has a significantinfluence on kinetic parameters, such as Tafel slope and exchange current density, and on corrosionand equilibrium potential. The authors proposed a “mixed” corrosion mechanism, with a generaldecrease of overvoltages and increase of exchange current density in the presence of AC. This effectcould be related to the local rise in temperature, associated with high AC densities, as reported in [16].In a recent work, Wu et al. [49] reported polarization curves of X70 steel tested with varying ACdensities (up to 100 A·m−2) in simulated seawater (Figure 14). AC shifts toward higher current densityin the polarization curves, promoting both the anodic and cathodic processes.

Materials 2020, 13, x FOR PEER REVIEW 14 of 21

The effect of spread resistance on AC corrosion has been also documented by Nielsen and Cohn [70] with the help of an electrical equivalent circuit analysis that represents the impedances existing between the pipe and remote earth. AC and DC sources impose a DC and AC voltage on the pipe: the simulated AC source is the HVTL (High Voltage Transmission Line), whereas the DC source represents the CP system. The authors consider static and dynamic elements. Static elements, namely spread resistance and charge transfer resistance, are defined as elements where impedance is frequency independent. Conversely, dynamic elements are frequency dependent: these include the double layer capacitance and diffusion elements. Because of the low impedance of the capacitance, the spread resistance is the dominant impedance element at 50–60 Hz frequency and plays a key role in controlling the AC corrosion process.

4.3. Effect of AC on Anodic and Cathodic Overvoltages

The increase of corrosion rate in the presence of AC has been explained by some authors using the effect of AC on anodic and cathodic overvoltages [39,46,49,65,71,72]. Goidanich et al. [39] investigated by means of laboratory tests the influence of AC on kinetic characteristics of carbon steel, galvanized steel, copper and zinc under different experimental conditions. Results showed that AC has a significant influence on kinetic parameters, such as Tafel slope and exchange current density, and on corrosion and equilibrium potential. The authors proposed a “mixed” corrosion mechanism, with a general decrease of overvoltages and increase of exchange current density in the presence of AC. This effect could be related to the local rise in temperature, associated with high AC densities, as reported in [16]. In a recent work, Wu et al. [49] reported polarization curves of X70 steel tested with varying AC densities (up to 100 A·m−2) in simulated seawater (Figure 14). AC shifts toward higher current density in the polarization curves, promoting both the anodic and cathodic processes.

Figure 14. Polarization curves on X70 steel tested at various AC densities in simulated seawater [49].

As discussed previously, some authors [43–47] proposed theoretical models based on the fundamental thermodynamic and kinetic laws of corrosion in order to investigate the effect of AC interference on DC potential and corrosion rate. Accordingly, the sensitivity of the corroding system is influenced by the ratio of the anodic-to-cathodic Tafel slope (r = βa/βc). The effect of r on DC potential variation has already been discussed in Section 2.6. Considering a corrosion system under activation control, Lalvani and Lin [45] proposed an analytical solution for the relationship between corrosion rate and voltage peak amplitude. In a revised model [46], the authors introduced the effect of the double layer capacity, without considering the resistance of the electrolyte. The model indicates that corrosion current increases with voltage peak for all values of r, while the potential shift depends strongly on the anodic and cathodic characteristic curves, i.e., on the ratio between the anodic and cathodic Tafel slope. Potentiodynamic polarization curves were obtained using the revised model;

Figure 14. Polarization curves on X70 steel tested at various AC densities in simulated seawater [49].

As discussed previously, some authors [43–47] proposed theoretical models based on thefundamental thermodynamic and kinetic laws of corrosion in order to investigate the effect ofAC interference on DC potential and corrosion rate. Accordingly, the sensitivity of the corrodingsystem is influenced by the ratio of the anodic-to-cathodic Tafel slope (r = βa/βc). The effect of r on DCpotential variation has already been discussed in Section 2.6. Considering a corrosion system underactivation control, Lalvani and Lin [45] proposed an analytical solution for the relationship betweencorrosion rate and voltage peak amplitude. In a revised model [46], the authors introduced the effectof the double layer capacity, without considering the resistance of the electrolyte. The model indicatesthat corrosion current increases with voltage peak for all values of r, while the potential shift dependsstrongly on the anodic and cathodic characteristic curves, i.e., on the ratio between the anodic andcathodic Tafel slope. Potentiodynamic polarization curves were obtained using the revised model;nevertheless, these approaches predict that corrosion current and corrosion potential are independentof the frequency of the AC signal, differently to what is observed.

Materials 2020, 13, 2158 16 of 22

The model was improved in 2008 [43,44]; the authors considered three elements in an electricalequivalent circuit of a metal subjected to an induced AC voltage: the polarization impedance, the doublelayer capacitance and the electrolyte resistance. The model shows that the corrosion current increasesas the frequency of the AC signal decreases (Figure 15a), in agreement with experimental observations,and by increasing the peak potential. Moreover, the model shows that corrosion current increases bydecreasing the DC corrosion potential. For instance, by decreasing the DC corrosion potential from−0.6 to −0.7 V vs. SCE at a peak potential of 1.25 V vs. SCE, the corrosion rate increases several ordersof magnitude (Figure 15b); an increase of DC corrosion potential from −0.2 to 0.0 V does not result ina further reduction of corrosion current [43].

Materials 2020, 13, x FOR PEER REVIEW 15 of 21

nevertheless, these approaches predict that corrosion current and corrosion potential are independent of the frequency of the AC signal, differently to what is observed.

The model was improved in 2008 [43,44]; the authors considered three elements in an electrical equivalent circuit of a metal subjected to an induced AC voltage: the polarization impedance, the double layer capacitance and the electrolyte resistance. The model shows that the corrosion current increases as the frequency of the AC signal decreases (Figure 15a), in agreement with experimental observations, and by increasing the peak potential. Moreover, the model shows that corrosion current increases by decreasing the DC corrosion potential. For instance, by decreasing the DC corrosion potential from −0.6 to −0.7 V vs. SCE at a peak potential of 1.25 V vs. SCE, the corrosion rate increases several orders of magnitude (Figure 15b); an increase of DC corrosion potential from −0.2 to 0.0 V does not result in a further reduction of corrosion current [43].

Recently, Ibrahim et al. [73–75] proposed a theoretical approach (Part 1, 2, 3) to evaluate the effect of double layer capacitance and electrolyte resistance on corrosion current density and potential shift. The authors stated that corrosion rate enhancement is due to the faradaic rectification as a consequence of the nonlinear current-potential relationship [73].

(a) (b)

Figure 15. A nonlinear model for corrosion of metals subjected to AC corrosion [43]: (a) dimensionless corrosion current vs. frequency, (b) dimensionless corrosion current vs. peak potential and DC corrosion potential. Legend: Ecorr,DC = corrosion potential in the absence of AC (V vs. SCE); Ep = AC peak potential (V vs. SCE); r = βa/βc.

4.4. Breakdown of the Passive Film and High-pH Corrosion

Recently, Brenna et al. [76,77] proposed a two-step AC corrosion mechanism. In the first step, AC causes the weakening of the passive film formed under CP condition on carbon steel, due to electromechanical stresses. Electrostriction appears to be a convincing explanation of the passive film breakdown mechanism, because of the presence of high alternating electric field (in the order of 106 V·cm−1 [78]) across the passive film [77]. The effect of AC on passive condition has been documented for carbon steel in alkaline solution or concrete and for stainless steel in neutral solution [79–85]: AC causes localized corrosion of passive metals with a decrease of corrosion resistance. Mechanical failure of the film can result from high electromechanical stresses (electrostriction pressure) generated by the presence of an electric field across the film and by the interfacial tension, which is not negligible as a result of the thin thickness of the oxide.

After film breakdown, high-pH chemical corrosion (i.e., potential independent) occurs in the overprotection condition because of the high cathodic current density supplied to the metal. According to the Pourbaix diagram of iron, high-pH corrosion can occur with formation of di-hypo ferrite ions (HFeO2−). This mechanism can also explain the unexpected corrosion of the metal below its equilibrium potential. Indeed, CP electrons are involved in the cathodic process, which depends on the potential assumed by the metal. Thus, below the equilibrium potential, the applied cathodic current makes electrons available to the metal; therefore, no anodic electrochemical reactions can take place. Consequently, if no oxidation reaction takes place, such as iron ion production, this can occur

Figure 15. A nonlinear model for corrosion of metals subjected to AC corrosion [43]: (a) dimensionlesscorrosion current vs. frequency, (b) dimensionless corrosion current vs. peak potential and DCcorrosion potential. Legend: Ecorr,DC = corrosion potential in the absence of AC (V vs. SCE); Ep = ACpeak potential (V vs. SCE); r = βa/βc.

Recently, Ibrahim et al. [73–75] proposed a theoretical approach (Part 1, 2, 3) to evaluate the effectof double layer capacitance and electrolyte resistance on corrosion current density and potential shift.The authors stated that corrosion rate enhancement is due to the faradaic rectification as a consequenceof the nonlinear current-potential relationship [73].

4.4. Breakdown of the Passive Film and High-pH Corrosion

Recently, Brenna et al. [76,77] proposed a two-step AC corrosion mechanism. In the firststep, AC causes the weakening of the passive film formed under CP condition on carbon steel,due to electromechanical stresses. Electrostriction appears to be a convincing explanation of thepassive film breakdown mechanism, because of the presence of high alternating electric field (in theorder of 106 V· cm−1 [78]) across the passive film [77]. The effect of AC on passive condition hasbeen documented for carbon steel in alkaline solution or concrete and for stainless steel in neutralsolution [79–85]: AC causes localized corrosion of passive metals with a decrease of corrosion resistance.Mechanical failure of the film can result from high electromechanical stresses (electrostriction pressure)generated by the presence of an electric field across the film and by the interfacial tension, which is notnegligible as a result of the thin thickness of the oxide.

After film breakdown, high-pH chemical corrosion (i.e., potential independent) occurs inthe overprotection condition because of the high cathodic current density supplied to the metal.According to the Pourbaix diagram of iron, high-pH corrosion can occur with formation of di-hypoferrite ions (HFeO2

−). This mechanism can also explain the unexpected corrosion of the metal below itsequilibrium potential. Indeed, CP electrons are involved in the cathodic process, which depends on thepotential assumed by the metal. Thus, below the equilibrium potential, the applied cathodic currentmakes electrons available to the metal; therefore, no anodic electrochemical reactions can take place.

Materials 2020, 13, 2158 17 of 22

Consequently, if no oxidation reaction takes place, such as iron ion production, this can occur onlythrough chemical corrosion, which is not potential dependent, or, in other words, is not influenced bythe electrons made available otherwise [86].

5. AC Corrosion Monitoring

According to ISO 18086 [4], AC voltage should be measured with respect to remote earth attest posts during the general and detailed assessment of CP effectiveness, as defined in ISO 15589-1.Additional measurements should be carried out at sites where AC corrosion risk is suspected, e.g.,areas where the soil resistivity is low (lower than 25 Ω·m), areas with highest AC interference levels,areas where AC corrosion has previously taken place and areas where local DC polarization conditionscan favor AC corrosion, as high levels of CP.

The measurement of AC and DC densities should be carried out by means of coupons or probesinstalled in the same soil or backfill as the pipeline itself. The measurements with respect to the criteriadefined in ISO 18086 have to be carried out on a 1 cm2 coupon surface area. Coupon or probe currentscan be measured by the voltage drop across a series resistor; the value of the resistance should besufficiently low to avoid disturbance of the system. For field measurements, a typical value is 10 Ω fora 1 cm2 coupon.

For corrosion rate measurements, weight loss measurements, perforation measurements orelectrical resistance (ER) measurements can be applied. Weight loss measurements require installationof pre-weighed coupons. After some time of operation (months to years), the coupon is excavated andweight loss rate is determined. The main drawback of this measurement is that the coupon providesno information until the excavation. Perforation measurements are made on special perforation probes:a signal is generated when the corrosion process has perforated the wall thickness of the coupon. In thiscase, information about the maximum (localized) corrosion depth is available without excavatingthe coupon; the primary disadvantage is that this information is not available until the coupon isperforated. Electrical resistance measurements require the installation of electrical resistance probes(ER probes). Corrosion is detected by the increase of the electrical resistance of the coupon whencorrosion progressively decreases its thickness [64,87,88].

6. Conclusions

This paper deals with a narrative literature review on AC corrosion assessment, protection criteriaand corrosion mechanisms for buried carbon steel structures in CP condition. Main conclusions can besummarized as follows:

• The assessment of AC corrosion likelihood should be based on the measurement of AC and DCrelated parameters. The AC interference level is evaluated by AC remote voltage and AC density,while the CP level is assessed by DC density and potential measurements;

• AC and DC densities should be measured by means of a corrosion coupon (1 cm2 area) connectedto the structure in CP condition; IR-free potential is considered more accurate than ON potential,because it does not contain the ohmic drop contribution in soil;

• There is general agreement that the DC potential of carbon steel in CP condition increases in thepresence of AC interference, although a negative shift is measured at small DC density. Conversely,in free corrosion condition, i.e., without CP, the potential decreases as the AC density increases;

• Overprotection (namely EIR-free < −1.2 V vs. CSE) is the most dangerous condition in thepresence of AC interference. At “high” CP levels, the maximum tolerable AC density is 30 A·m−2.Below 1 A·m−2 DC density, the AC corrosion likelihood decreases. Nevertheless, some doubts arerevealed regarding the inexistence of the criterion reported in the ISO 18086 standard of an ACdensity threshold at “low” CP condition (iDC < 1 A·m−2);

• The higher AC corrosion likelihood at high CP levels could be explained by a corrosion mechanismthat involves both the AC and DC levels:

Materials 2020, 13, 2158 18 of 22

At high DC density, a strong alkalization of the electrolyte close to the coating defect occurswith formation of a passive film and deposits (as a calcareous deposit) on carbon steel.Soil chemical composition, pH and spread resistance at the coating defects seem to havea crucial role, controlling the local AC and DC densities;

AC interference provokes a weakening of the passive condition due to an effect on anodicand cathodic overvoltages; moreover, the scale formed in CP condition is not stable in thepresence of AC due to potential oscillations that could break the protective layer;

High-pH corrosion occurs with localized corrosion attacks; chemical corrosion (i.e., potentialindependent) with formation of di-hypo ferrite ions (HFeO2

−) is a possible explanation forthe occurrence of corrosion at low potentials.

Author Contributions: A.B. conceptualized and prepared the manuscript; M.O. and S.B. contributed to literatureanalysis and to the organization of the manuscript. All authors have read and agreed to the published version ofthe manuscript.

Funding: This research received no external funding.

Conflicts of Interest: The authors declare no conflicts of interest.

References

1. General Principles of Cathodic Protection of Buried or Immersed Onshore Metallic Structures; EN 12954;British Standards Institution: London, UK, 29 August 2019.

2. Petroleum, Petrochemical and Natural Gas Industries. Cathodic Protection of Pipeline Systems. Part 1: On-land Pipelines;ISO 15589-1; ISO - International Organization for Standardization: Geneva, Switzerland, 1 March 2015.

3. CIGRE Working Group 36.02. Guide on the Influence of High Voltage AC Power Systems on Metallic Pipelines;CIGRE Technical Brochure no. 095; CIGRE: Paris, France, 1995.

4. Corrosion of Metals and Alloys. Determination of AC Corrosion. Protection Criteria; ISO 18086; ISO - InternationalOrganization for Standardization: Geneva, Switzerland, 2019.

5. Effects of Electromagnetic Interference on Pipelines Caused by High Voltage a.c. Electric Traction Systems and/or HighVoltage a.c. Power Supply Systems; EN 50443; CEN-CENELEC - European Committee for ElectrotechnicalStandardization: Brussels, Belgium, 1 January 2011.

6. IEEE. IEEE Guide for Safety in AC Substation Grounding; IEEE Std. 80-2000; IEEE - The Institute of Electricaland Electronics Engineers, Inc.: New York, NY, USA, 2000.

7. Mitigation of Alternating Current and Lightning Effects on Metallic Structures and Corrosion Control Systems;NACE SP0177; NACE International: Houston, TX, USA, 22 June 2007.

8. Tang, D.; Du, Y.; Lu, M.; Chen, S.; Jiang, Z.; Dong, L. Study on location of reference electrode for measurementof induced alternating current voltage on pipeline. Int. Trans. Electr. Energy Syst. 2013, 25, 99–119. [CrossRef]

9. NACE International Task Group 327. AC corrosion State-of-the-Art: Corrosion Rate, Mechanism, and MitigationRequirements; NACE Technical Committee Report 35110; NACE International: Houston, TX, USA, 2010.

10. Gummow, R.A.; Segall, S.M.; Fieltsch, W. Pipeline ac mitigation misconceptions. In Proceedings of theNorthern Area Western Conference, Calgary, AB, Canada, 15–18 February 2010; p. 14.

11. Panossian, Z.; Filho, S.E.A.; De Almeida, N.L.; Filho, M.L.P.; De L. Silva, D.; Laurino, E.W.; De L. Oliver, J.H.;De S. Pimenta, G.; De C. Albertini, J.Á. Effect of alternating current by high power lines voltage and electrictransmission systems in pipelines corrosion. In Proceedings of the Corrosion 2009 Conference & Expo,Atlanta, GA, USA, 22–26 March 2009; p. 41.

12. Wakelin, R.G.; Gummow, R.A.; Segall, S.M. AC corrosion—Case histories, test procedures, & mitigation.In Proceedings of the Corrosion/98, San Diego, CA, USA, 22–27 March 1998; p. 14.

13. Ouadah, M.; Touhami, O.; Ibtiouen, R.; Zergoug, M. Method for diagnosis of the effect of AC on the X70pipeline due to an inductive coupling caused by HVPL. IET Sci. Meas. Technol. 2017, 11, 766–772. [CrossRef]

14. Williams, J.F. Corrosion of metals under the influence of alternating current. Mater. Prot. 1966, 5, 52–53.15. Pookote, S.R.; Chin, D.-T. Effect of alternating current on the underground corrosion of steels.

Mater. Performance 1978, 17, 9–15.

Materials 2020, 13, 2158 19 of 22

16. Goidanich, S.; Lazzari, L.; Ormellese, M. AC corrosion. Part 2: Parameters influencing corrosion rate.Corros. Sci. 2010, 52, 916–922. [CrossRef]

17. Fu, A.; Cheng, Y.F. Effects of alternating current on corrosion of a coated pipeline steel in a chloride-containingcarbonate/bicarbonate solution. Corros. Sci. 2010, 52, 612–619. [CrossRef]

18. Gummow, R.A.; Wakelin, R.G.; Segall, S.M. AC corrosion—A new challenge to pipeline integrity.In Proceedings of the Corrosion/98, San Diego, CA, USA, 22–27 March 1998; p. 18.

19. Pourbaix, A.; Carpentiers, P.; Gregoor, R. Detection and assessment of AC induced corrosion.Mater. Performance 2000, 39, 34–37.

20. Yunovich, M.; Thompson, N.G. AC corrosion: Corrosion rate and mitigation requirements. In Proceedingsof the Corrosion/2004, New Orleans, LA, USA, 22 March–1 April 2004; p. 18.

21. Ragault, I. AC corrosion induced by V.H.V. electrical lines on polyethylene coated steel gas pipelines.In Proceedings of the Corrosion/98, San Diego, CA, USA, 22–27 March 1998; p. 14.

22. Büchler, M.; Schöneich, H.-G. Investigation of Alternating Current Corrosion of Cathodically ProtectedPipelines: Development of a Detection Method, Mitigation Measures, and a Model for the Mechanism.Corrosion 2009, 65, 578–586. [CrossRef]

23. Ormellese, M.; Lazzari, L.; Brenna, A.; Trombetta, A. Proposal of CP criterion in the presence of AC-interference.In Proceedings of the Corrosion 2010 Conference & Expo, San Antonio, TX, USA, 14–18 March 2010; p. 17.

24. He, X.; Jiang, G.; Qiu, Y.; Zhang, G.; Jin, X.; Xiang, Z.; Zhang, Z.; Tang, H. Study of criterion for assuringthe effectiveness of cathodic protection of buried steel pipelines being interfered with alternative current.Mater. Corros. 2011, 63, 534–543. [CrossRef]

25. Ding, H.; Li, J.; Wang, S.; Yang, Y. Experimental study on stray current corrosion of coated pipeline steel.J. Nat. Gas Sci. Eng. 2015, 27, 1555–1561. [CrossRef]

26. Ormellese, M.; Lazzari, L.; Brenna, A. AC corrosion of cathodically protected buried pipelines: Criticalinterference values and protection criteria. In Proceedings of the Corrosion 2015 Conference & Expo, Dallas,TX, USA, 15–19 March 2015; p. 11.

27. Xiao, Y.; Du, Y.; Tang, D.; Ou, L.; Sun, H.; Lu, Y. Study on the influence of environmental factors on ACcorrosion behavior and its mechanism. Mater. Corros. 2017, 69, 601–613. [CrossRef]

28. Fernandes, S.Z.; Mehendale, S.G.; Venkatachalam, S. Influence of frequency of alternating current on theelectrochemical dissolution of mild steel and nickel. J. Appl. Electrochem. 1980, 10, 649–654. [CrossRef]

29. Lalvani, S.; Zhang, G. The corrosion of carbon steel in a chloride environment due to periodic voltagemodulation: Part I. Corros. Sci. 1995, 37, 1567–1582. [CrossRef]

30. Qiu, W.W.; Pagano, M.; Zhang, G.; Lalvani, S. A periodic voltage modulation effect on the corrosion oFCu-Ni Alloy. Corros. Sci. 1995, 37, 97–110. [CrossRef]

31. Lalvani, S.B.; Kang, J.-C.; Mandich, N.V. The corrosion of Cu-Ni alloy in a chloride solution subjected toperiodic voltage modulation: Part I. Corros. Sci. 1998, 40, 69–89. [CrossRef]

32. Lalvani, S.B.; Kang, J.-C.; Mandich, N.V. The corrosion of Cu-Ni alloy in a chloride solution subjected toperiodic voltage modulation: Part II. Corros. Sci. 1998, 40, 201–214. [CrossRef]

33. Song, H.; Kim, Y.; Lee, S.; Kho, Y.; Park, Y. Competition of AC and DC current in AC corrosion under cathodicprotection. In Proceedings of the Corrosion/2002, Denver, CO, USA, 7–11 April 2002; p. 12.

34. Guo, Y.-B.; Liu, C.; Wang, D.-G.; Liu, S.-H. Effects of alternating current interference on corrosion of X60pipeline steel. Pet. Sci. 2015, 12, 316–324. [CrossRef]

35. Guo, Y.; Meng, T.; Wang, D.; Tan, H.; He, R. Experimental research on the corrosion of X series pipeline steelsunder alternating current interference. Eng. Fail. Anal. 2017, 78, 87–98. [CrossRef]

36. Yunovich, M.; Thompson, N.G. AC Corrosion: Mechanism and Proposed Model. In Proceedings of the 2004International Pipeline Conference, Calgary, AB, Canada, 4–8 October 2004; Volumes 1–3, pp. 183–195.

37. Büchler, M.; Voûte, C.-H.; Schöneich, H.-G.; Stalder, F. Characteristics of potential measurements in the fieldof ac corrosion. In Proceedings of the CeoCor International Congress and Technical Exhibition, GiardiniNaxos, Italy, 13–16 May 2003.

38. Bolzoni, F.; Goidanich, S.; Lazzari, L.; Ormellese, M. Laboratory test results of AC interference on polarizedsteel. In Proceedings of the Corrosion/2003, San Diego, CA, USA, 16–20 April 2003; p. 13.

39. Goidanich, S.; Lazzari, L.; Ormellese, M. AC corrosion—Part 1: Effects on overpotentials of anodic andcathodic processes. Corros. Sci. 2010, 52, 491–497. [CrossRef]

Materials 2020, 13, 2158 20 of 22

40. Bolzoni, F.; Goidanich, S.; Lazzari, L.; Ormellese, M.; Pedeferri, M. Laboratory testing on the influenceof alternated current on steel corrosion. In Proceedings of the Corrosion/2004, New Orleans, LA, USA,22 March–1 April 2004; p. 11.

41. Li, Y. Effects of Stray AC Interference on Corrosion Behavior of X70 Pipeline Steel in a Simulated Marine SoilSolution. Int. J. Electrochem. Sci. 2017, 12, 1829–1845. [CrossRef]

42. Wang, X. Corrosion Behavior of X70 and X80 Pipeline Steels in Simulated Soil Solution. Int. J. Electrochem. Sci.2018, 13, 6436–6450. [CrossRef]

43. Zhang, R.; Vairavanathan, P.R.; Lalvani, S. Perturbation method analysis of AC-induced corrosion. Corros. Sci.2008, 50, 1664–1671. [CrossRef]

44. Xiao, H.; Lalvani, S.B. A Linear Model of Alternating Voltage-Induced Corrosion. J. Electrochem. Soc.2008, 155, C69. [CrossRef]

45. Lalvani, S.; Lin, X. A theoretical approach for predicting AC-induced corrosion. Corros. Sci. 1994, 36, 1039–1046.[CrossRef]

46. Lalvani, S.; Lin, X. A revised model for predicting corrosion of materials induced by alternating voltages.Corros. Sci. 1996, 38, 1709–1719. [CrossRef]

47. Bosch, R.; Bogaerts, W. A theoretical study of AC-induced corrosion considering diffusion phenomena.Corros. Sci. 1998, 40, 323–336. [CrossRef]

48. Xu, L.; Su, X.; Yin, Z.; Tang, Y.; Cheng, Y.F. Development of a real-time AC/DC data acquisition technique forstudies of AC corrosion of pipelines. Corros. Sci. 2012, 61, 215–223. [CrossRef]

49. Wu, W.; Pan, Y.; Liu, Z.; Du, C.; Li, X. Electrochemical and Stress Corrosion Mechanism of Submarine Pipeline inSimulated Seawater in Presence of Different Alternating Current Densities. Materials 2018, 11, 1074. [CrossRef]

50. Ormellese, M.; Lazzari, L.; Goidanich, S.; Sesia, V. CP criteria assessment in the presence of AC interference.In Proceedings of the Corrosion 2008 Conference & Expo, New Orleans, LA, USA, 16–20 March 2008; p. 10.

51. Xu, L.; Su, X.; Cheng, Y.F. Effect of alternating current on cathodic protection on pipelines. Corros. Sci.2013, 66, 263–268. [CrossRef]

52. Xu, L.; Cheng, Y.F. A Real-Time AC/DC Measurement Technique for Investigation of AC Corrosion ofPipelines and Its Effect on the Cathodic Protection Effectiveness. In Proceedings of the Corrosion 2013Conference & Expo, Orlando, FL, USA, 17–21 March 2013; p. 10.

53. Kuang, D.; Cheng, Y.F. Effect of alternating current interference on coating disbondment and cathodicprotection shielding on pipelines. Corros. Eng. Sci. Technol. 2015, 50, 211–217. [CrossRef]

54. Kuang, D.; Cheng, Y.F. Effects of alternating current interference on cathodic protection potential and itseffectiveness for corrosion protection of pipelines. Corros. Eng. Sci. Technol. 2016, 52, 1–7. [CrossRef]

55. Wang, L.; Cheng, L.; Li, J.; Zhu, Z.; Bai, S.; Cui, Z. Combined Effect of Alternating Current Interference andCathodic Protection on Pitting Corrosion and Stress Corrosion Cracking Behavior of X70 Pipeline Steel inNear-Neutral pH Environment. Materials 2018, 11, 465. [CrossRef] [PubMed]

56. Hosokawa, Y.; Kajiyama, F.; Nakamura, Y. New CP criteria for elimination of the risks of ac corrosion andoverprotection on cathodically protected pipelines. In Proceedings of the Corrosion/2002, Denver, CO, USA,7–11 April 2002; p. 12.

57. Hosokawa, Y.; Kajiyama, F. New CP maintenance concept for buried steel pipelines—Current density-basedCP criteria, and on-line surveillance system for CP rectifiers. In Proceedings of the Corrosion/2004,New Orleans, LA, USA, 22 March–1 April 2004; p. 15.

58. Hosokawa, Y.; Kajiyama, F. Case studies on the assessment of AC and DC interference using steel couponswith respect to current density CP criteria. In Proceedings of the Corrosion 2006 Conference & Expo, Orlando,FL, USA, 10–14 September 2006; p. 15.

59. Kajiyama, F.; Nakamura, Y. Development of an advanced instrumentation for assessing the AC corrosionrisk of buried pipelines. In Proceedings of the Corrosion 2010 Conference & Expo, San Antonio, TX, USA,14–18 March 2010; p. 13.

60. Fu, A.Q.; Cheng, Y.F. Effect of alternating current on corrosion and effectiveness of cathodic protection ofpipelines. Can. Met. Q. 2012, 51, 81–90. [CrossRef]

61. Büchler, M. Alternating current corrosion of cathodically protected pipelines: Discussion of the involvedprocesses and their consequences on the critical interference values. Mater. Corros. 2012, 63, 1181–1187.[CrossRef]

Materials 2020, 13, 2158 21 of 22

62. Junker, A.; Nielsen, L.V.; Heinrich, C.; Møller, P. Laboratory and field investigation of the effect of thechemical environment on AC corrosion. In Proceedings of the Corrosion 2018 Conference & Expo, Phoenix,AZ, USA, 15–19 April 2018; p. 15.

63. Nielsen, L.V. AC Corrosion. In Oil and Gas Pipelines; John Wiley & Sons: Hoboken, NJ, USA, 2015; pp. 363–386.64. Nielsen, L.V.; Nielsen, K.V.; Baumgarten, B.; Breuning-Madsen, H.; Cohn, P.; Rosenberg, H. AC-induced

corrosion in pipelines: Detection, characterization, and mitigation. In Proceedings of the Corrosion/2004,New Orleans, LA, USA, 22 March–1 April 2004; p. 16.

65. Nielsen, L.V. Role of alkalization in AC induced corrosion of pipelines and consequences hereof in relationto CP requirements. In Proceedings of the Corrosion/2005, Houston, TX, USA, 3–7 April 2005; p. 11.

66. Nielsen, L.V.; Baumgarten, B.; Cohn, P. On-site measurements of AC induced corrosion: Effect of AC and DCparameters—A report from the Danish activities. In Proceedings of the CeoCor International Congress andTechnical Exhibition, Dresden, Germany, 15–16 June 2004.

67. Junker, A.; Møller, P.; Nielsen, L.V. Effect of chemical environment and pH on AC corrosion of cathodicallyprotected structures. In Proceedings of the Corrosion 2017 Conference & Expo, New Orleans, LA, USA,26–30 March 2017; p. 14.

68. Junker, A.; Belmonte, L.J.; Kioupis, N.; Nielsen, L.V.; Møller, P. Investigation of stone-hard-soil formationfrom AC corrosion of cathodically protected pipeline. Mater. Corros. 2018, 69, 1170–1179. [CrossRef]

69. Santana-Diaz, E.; Nielsen, L.V.; Junker-Holst, A. A critical review of parameters for meaningful AC corrosionmodelling. In Proceedings of the Corrosion 2018 Conference & Expo, Phoenix, AZ, USA, 15–19 April 2018; p. 15.

70. Nielsen, L.V.; Cohn, P. AC corrosion and electrical equivalent diagrams. In Proceedings of the CeoCorInternational Congress and Technical Exhibition, Bruxelles, Belgium, 9–12 May 2000.

71. Wang, X.; Wang, Z.; Chen, Y.; Song, X.; Xu, C. Research on the Corrosion Behavior of X70 Pipeline SteelUnder Coupling Effect of AC + DC and Stress. J. Mater. Eng. Perform. 2019, 28, 1958–1968. [CrossRef]

72. Cui, Y.; Shen, T.; Ding, Q. Study on the Influence of AC Stray Current on X80 Steel under Stripped Coatingby Electrochemical Method. Int. J. Corros. 2019, 2019, 1–8. [CrossRef]

73. Ibrahim, I.; Tribollet, B.; Takenouti, H.; Meyer, M. AC-Induced Corrosion of Underground Steel Pipelines.Faradaic Rectification under Cathodic Protection: I. Theoretical Approach with Negligible ElectrolyteResistance. J. Braz. Chem. Soc. 2014, 26, 196–208. [CrossRef]

74. Ibrahim, I.; Meyer, M.; Takenouti, H.; Tribollet, B. AC Induced Corrosion of Underground Steel Pipelines.Faradaic Rectification under Cathodic Protection: II. Theoretical Approach with Electrolyte Resistance andDouble Layer Capacitance for Bi-Tafelian Corrosion Mechanism. J. Braz. Chem. Soc. 2015, 27, 605–615.[CrossRef]

75. Ibrahim, I.; Meyer, M.; Takenouti, H.; Tribollet, B. AC Induced Corrosion of Underground Steel Pipelines underCathodic Protection: III. Theoretical Approach with Electrolyte Resistance and Double Layer Capacitance forMixed Corrosion Kinetics. J. Braz. Chem. Soc. 2017, 28, 1483–1493. [CrossRef]

76. Brenna, A.; Diamanti, M.V.; Lazzari, L.; Ormellese, M. A proposal of AC corrosion mechanism in cathodicprotection. In Proceedings of the 2011 NSTI Nanotechnology Conference and Expo, Boston, MA, USA, 13–16June 2011; pp. 553–556.

77. Brenna, A.; Ormellese, M.; Lazzari, L. Electro-mechanical breakdown mechanism of passive film in AC-relatedcorrosion of carbon steel under cathodic protection condition. Corrosion 2016, 72, 1055–1063. [CrossRef]

78. Sato, N. A theory for breakdown of anodic oxide films on metals. Electrochimica Acta 1971, 16, 1683–1692.[CrossRef]

79. Bertolini, L.; Carsana, M.; Pedeferri, P. Corrosion behaviour of steel in concrete in the presence of straycurrent. Corros. Sci. 2007, 49, 1056–1068. [CrossRef]

80. Ormellese, M.; Brenna, A.; Lazzari, L. Effects of AC-interference on passive metals corrosion. In Proceedingsof the Corrosion 2011 Conference & Expo, Houston, TX, USA, 13–17 March 2011; p. 12.

81. Zhu, M.; Du, C.-W.; Li, X.; Liu, Z.-Y.; Wu, X.-G. Effect of AC on corrosion behavior of X80 pipeline steel inhigh pH solution. Mater. Corros. 2014, 66, 486–493. [CrossRef]

82. Brenna, A.; Beretta, S.; Bolzoni, F.M.; Pedeferri, M.; Ormellese, M. Effects of AC-interference onchloride-induced corrosion of reinforced concrete. Constr. Build. Mater. 2017, 137, 76–84. [CrossRef]

83. Zhu, M. Corrosion Behavior of X65 and X80 Pipeline Steels under AC Interference Condition in High pHSolution. Int. J. Electrochem. Sci. 2018, 13, 10669–10678. [CrossRef]

Materials 2020, 13, 2158 22 of 22

84. Zhu, M. Synergistic Effect of AC and Cl- on Stress Corrosion Cracking Behavior of X80 Pipeline Steel inAlkaline Environment. Int. J. Electrochem. Sci. 2018, 13, 10527–10538. [CrossRef]

85. Wang, H.; Du, C.; Liu, Z.; Wang, L.; Ding, D. Effect of Alternating Current on the Cathodic Protection andInterface Structure of X80 Steel. Materials 2017, 10, 851. [CrossRef]

86. Drazic, D.; Popic, J. Anomalous dissolution of metals and chemical corrosion. J. Serbian Chem. Soc. 2005, 70, 489–511.[CrossRef]

87. Nielsen, L.V.; Nielsen, K.V. Differential ER-technology for measuring degree of accumulated corrosion as well asinstant corrosion rate. In Proceedings of the Corrosion/2003, San Diego, CA, USA, 16–20 April 2003; p. 13.

88. Nielsen, L.V.; Galsgaard, F. Sensor technology for on-line monitoring of ac-induced corrosion along pipelines.In Proceedings of the Corrosion/2005, Houston, TX, USA, 3–7 April 2005; p. 12.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open accessarticle distributed under the terms and conditions of the Creative Commons Attribution(CC BY) license (http://creativecommons.org/licenses/by/4.0/).